Physica Medica: European Journal of Medical Physics
Volume 27, Issue 1 , Pages 52-57, January 2011

Physical characterization and comparison of two commercially available micro-MLCs

  • Tarun K. Podder

      Affiliations

    • Department of Radiation Oncology, Kimmel Cancer Center (NCI-designated), Jefferson Medical College, Thomas Jefferson University Hospital, Philadelphia, PA 19107, USA
    • Corresponding Author InformationCorresponding author. Tel.: +1 215 955 9290.
    • ,
  • Greg Bednarz

      Affiliations

    • Department of Radiation Oncology, UPMC Cancer Center, University of Pittsburgh Medical Center, Pittsburgh, PA 15232, USA
    • ,
  • Yan Yu

      Affiliations

    • Department of Radiation Oncology, Kimmel Cancer Center (NCI-designated), Jefferson Medical College, Thomas Jefferson University Hospital, Philadelphia, PA 19107, USA
    • ,
  • James M. Galvin

      Affiliations

    • Department of Radiation Oncology, Kimmel Cancer Center (NCI-designated), Jefferson Medical College, Thomas Jefferson University Hospital, Philadelphia, PA 19107, USA

Received 3 April 2009; received in revised form 28 December 2009; accepted 27 January 2010. published online 04 March 2010.

Article Outline

Abstract 

In this study, the physical characteristics (penumbra width variation with the source size and shape, interleaf leakage, transmission through the leaves, and the tongue-and-groove effect) of two linear accelerators (BrainLAB's Novalis and Elekta's Synergy-S Beam Modulator) have been investigated. For similar square fields (about 4.5cm×4.5cm) with source-to-surface/skin-distance (SSD) ranging from 90cm to 115cm and measurements taken at the depth of Dmax=1.5cm for 6MV photon beam. The Novalis MLC has penumbra width of 2.4±0.11mm–2.8±0.11mm at the leaf-end and 2.2±0.1mm–2.7±0.1mm on the leaf-side; and those for the Synergy-S MLC are 4.4±0.17mm–5.2±0.2mm and 3.0±0.12mm–3.5±0.12mm. Upon rotating the Synergy-S collimator by 90° (i.e., shifting the leaf movement to the gun–target direction), significant reduction of the leaf-end penumbra width (17%) and increase of leaf-side penumbra width (28%) suggest an elliptical shape of the radiation source spot. Similar rotation of the collimator yielded reduction of the penumbras on both leaf-end (34%) and leaf-side (28%) for Novalis, indicating that the Novalis has a more symmetric source size. For all the field sizes and settings, BrainLAB's Novalis μMLC produce a smaller penumbra for simple square fields compared to the Elekta's Synergy-S. However, this difference became less pronounced for leaf-side penumbra and also for circular fields. The tongue-and-groove effect of the Novalis (23±0.9%) is slightly smaller than that of the Synergy-S (25±1%); while the interleaf leakage and leakage directly through leaves for Synergy-S (1.6±0.07% & 0.9±0.04%) are lower than that of Novalis (2±0.08% & 1.3±0.05%).

Keywords: Micromulti-leaf collimator, Penumbra, Interleaf leakage, Tongue-and-groove effect, Focal spot, IMRT

 

Back to Article Outline

Introduction 

Most commonly used integrated MLC ranges from 10mm to 3mm width as projected to the machine isocenter (normally 100cm source-to-axis (SAD)). A smaller MLC leaf width is a desirable characteristic for obtaining improved field conformality around irregularly shaped targets. There are various studies in the literature related to comparison of physical characteristics and dosimetric impact of different MLC [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [14], [15], [16]; interleaf transmission [1], optimum leaf width [2], basic applications [3], comparison of micro-MLC [4], [5], [9], [11], characterization [6], [7], [8], [10], [16] and dosimetric consideration [12], [14], [15].

Using EDR2 film (Kodak, Rochester, NY), Clark et al. [4] reported the penumbra characteristics of the BrainLAB's Novalis and Varian's Millenium MLC, which have central leaf widths of 3mm and 5mm, respectively. The 3mm MLC showed 80%–20% penumbra width of 2.4±0.1mm in both the directions perpendicular and parallel to the leaf motion. The 5mm MLC showed comparable penumbra width of 2.5±0.1mm in the direction perpendicular to leaf motion and 3.2±0.1mm in the direction parallel to leaf motion. Chern et al. [5] reported that the 3mm μMLC of BrainLAB provided better target conformity and greater normal tissue sparing than the 5mm MLC in stereotactic radiosurgery using dynamic conformal arcs. García-Garduño et al. [8] have recently reported penumbra (2.26±0.11mm along leaf side and 2.31±0.11mm along lead end), transmission (0.93±0.05%), and leakage (1.08±0.08%) of m3-μMLC of BrainLAB for 6MV photon beam study using EBT film.

Patel et al. [14] investigated the dosimetric characteristics (leaf bank setup, penumbra width, leaf positioning reproducibility, interleaf leakage and leaf transmission) of a production pilot mMLC (Elekta Beam Modulator™, Elekta Oncology Systems, Crawley, UK) having 4 mm leaf width at isocenter. The penumbra values for leaf ends were measured to be between 4.2 mm and 4.8 mm for various large rectangular fields.

In this paper, the physical characteristics of μMLC from two linacs, one is the BrainLAB's Novalis and the other is Elekta's Synergy-S Beam Modulator, have been evaluated. The Novalis has three types of leaves projecting to 3mm, 4.5mm and 5.5mm at isocenter, whereas the Synergy-S uses the Elekta Beam Modulator MLC with a 4mm leaf width at isocenter. The Novalis machine is in clinical use for a long time; whereas, the Synergy-S is relatively new in the clinic (about 4 years). However, so far the physical characterization of the clinical Synergy-S has not been extensively evaluated in clinical settings. Both the linacs can be used for radiosurgery applications that require small field shaping with tight tumor margin, where the beam conformality to the PTV and narrow penumbra for any configuration of the collimator is highly desirable. The difference in spot size and shape may have significant effects on the variation of penumbra at different configuration of the collimators for both these machines. This aspect has not been investigated in any of the previous studies. Thus, it is important to investigate the physical characteristics especially, penumbra width for different collimator orientations. Additionally, interleaf leakage, direct leakage through the leaves, and the tongue-and-groove effect for these two machines which have comparable MLCs have been studied.

Back to Article Outline

Materials and methods 

System overview 

BrainLAB's Novalis linac has 26 leaf pairs mounted at a distance of about 44cm from the radiation source, i.e., the focal spot. The maximum field size at the isocentric plane that can be used is 9.6cm×10.2cm. The leaf width (mutually perpendicular to both leaf movement direction and the beam central axis) at 100cm is 3mm for central 14 leaf pairs, 4.5mm for the next 3 leaf pairs on each side of the central leaves and 5.5mm for the remaining 3 leaf pairs on each side of the central leaves. The physical leaf thickness is 6.4cm and the leaf tip is rounded. There are two sets of backup jaws that work in combination with the μMLC leaves for enhancing beam attenuation.

Elekta's Synergy-S Beam Modulator has a total of 40 pairs of leaves with a 4mm width measured at 100cm and the physical leaf thickness is 7.5cm, and the leaf tip is rounded. The MLC is installed at a distance of 39cm from the source of radiation. The maximum field at the isocentric plane is 21cm×16cm. In this study, both linacs have been characterized for 6MV photon energy for same or similar field sizes. The Synergy-S Beam Modulator system does not use a backup jaw system.

Penumbra measurement 

In this study, the physical penumbra characteristics of the Elekta's Synergy-S μMLC and the BrainLAB's Novalis μMLC were investigated by exposing Kodak EDR2 (Eastman Kodak Inc., Rochester, NY) films at different configurations, i.e., various field sizes and shapes (SSD=100cm and depth=Dmax are kept constant), SSDs (depth=Dmax), and depths of measurement (constant SSD=100cm). For evaluating the effects of SSD, a range of SSD (90cm to 115cm) was considered while the films were placed at the depth of Dmax=1.5cm. The EDR2 film was calibrated by generating the Hurter and Driffield (H&D) curve, plotting dose vs. optical density following the guideline of the AAPM Task Group 69 [13]. Exposed films were scanned using VIDAR's DiagnosticPRO Advantage film digitizer (VIDAR Systems Corp., Herndon, VA). To investigate the effect of the shape of the source, i.e., focal spot size, two sets of measurements were taken by orienting the collimator at 0° and at 90° with the gantry at 0°. Square and circular fields ranging from 3cm to 10cm were used. The SSD=100cm was used for studying the effect of field size and shape. The effects of measurement depth were investigated for a depth ranging from 1.5cm to 15cm while the SAD was kept constant at 100cm. The effective penumbra information was expressed as lateral distance of 80%–20% isodose lines [4], [7], [8], [9], [14], which is considered more relevant for clinical applications. For all the film measurements three sets of films were exposed for each of the settings and 10 measurements at different location were taken on the irradiated films for calculating mean and standard deviation (SD). The SD was incorporated into the mean value to show the measurement uncertainties as mean±SD.

Tongue-and-groove effect 

The tongue and grove (T&G) effects were measured by pulling in a set of leaves to block the radiation beam while approximately 100cGy (100MU) dose was delivered. Then this set of leaves was retracted and the alternate set of leaves was brought into the field to block the beam and another 100MU was delivered. The T&G effects as well as the leakage measurement were performed using the same EDR2 films as mentioned above in Section Penumbra measurement.

Leakage measurement 

The leakage measurements were performed by exposing radiographic films while the radiation beam was blocked using the MLC for both Novalis and Synergy-S. For Novalis, the jaws were parked just outside of the maximum field (9.6cm×10.2cm), which was covered by the MLC only. The films were exposed to about 5000cGy (5000MU) placing at 100cm SAD through 1.5cm solid water buildup. The percentage leakage was defined for characterizing the interleaf leakage and leakage through leaves.

Back to Article Outline

Results and discussions 

Penumbra 

The results of the penumbra measurements for 6MV photon beam are presented in Fig. 1 through Fig. 3. In Fig. 1a, the penumbras of Synergy-S and Novalis for both leaf-end (along the direction of motion of the leaves) and leaf-side (in the direction perpendicular to the motion of the leaves) for SSDs from 90cm to 115cm at constant depth=Dmax and square field of 4.8cm2 for Synergy-S and 4.2cm2 for Novalis at isocenter of 100cm, with the collimator at 0° orientation have been shown. It is to be noted that the square field of 4.2cm2 for Novalis was created by using the central 14 pairs of leaves of 3mm width, while the 4.8cm2 field of Synergy-S was made by using 12 pairs of central leaves of 4mm width. The leaf-end penumbra width for SSD 90–115cm is in the range from 4.4±0.17mm to 5.2±0.2mm for Synergy-S (top curve) and it is much larger compared to the leaf-side penumbra (range 3.0±0.12mm-3.5±0.12mm) for this linac. On the contrary, both the leaf-end (range 2.4±0.11mm–2.8±0.11mm) and the leaf-side (range 2.2±0.1mm–2.7±0.1mm) penumbra widths for the Novalis (bottom two curves) are close to each other as well as much smaller than respective penumbras of the Synergy-S. The average difference of penumbra at the leaf-side and at the leaf-end is about 50% for the Elekta's Synergy-S Beam Modulator, while this difference is about 6% for the Novalis' μMLC. These differences between the Synergy-S and the Novalis penumbras could be mainly related to: (1) the source to diaphragm/collimator distance (SDD=44cm for Novalis and SDD=39cm for Synergy-S), (2) the source size and shape. Shorter SDD, and larger source size contribute to increase in the penumbra of Synergy-S (the contribution from field size difference of 0.6cm, i.e., 4.8cm vs. 4.2cm is relatively small). The closeness of the leaf-end and leaf-side penumbras of the Novalis can be attributed to the shape and size of the focal spot (approximately circular, smaller) and the design of the leaf-end and the leaf-side (tongue-groove).

  • View full-size image.
  • Figure 1 

    a) Penumbra width of Elekta's Synergy-S and BrainLAB's Novalis at Dmax=1.5cm of 6MV photon for SSD setup while the collimator was at 0-deg. (b) Penumbra of Elekta's Synergy-S and BrainLAB's Novalis at Dmax of 6MV photon for SSD setup. Collimator at 90-deg.

The collimator system was rotated to the 90° position to determine if elongation of the radiation spot size explained the difference in the leaf end and leaf-side penumbra widths for the Elekta collimator. It was interesting to observe that after collimator rotation, the size of the two penumbra widths (the leaf-end and the leaf-side) for the Synergy-S moved closer to an average position relative to those seen in Fig. 1a (top two curves in Fig. 1b). The leaf-end penumbra decreased by 17% and leaf-side penumbra increased by 28% (see Table 1 below). This suggests that the shape of the X-ray spot is elliptical with the minor axis along the gun–target direct of the linac. On the other hand, similar rotation of the collimator yielded significant decrease of the penumbra widths on both leaf-end (34%) and leaf-side (28%) for Novalis (bottom two curves in Fig. 1b). This indicates that the Novalis has approximately circular source as compared to that of the Synergy-S.

Table 1. Change in penumbra width due to change in collimator orientation.
LinacLeaf-end average penumbra at Dmax for 90cm–115cm SSDLeaf-side average penumbra at Dmax for 90cm–115cm SSD
Collimator at 0-degCollimator at 90-degChangeCollimator at 0-degCollimator at 90-degChange
Synergy-S4.8mm4.0mm−17 % ↓3.2mm4.1mm28 % ↑
Novalis2.6mm1.7mm−34 % ↓2.5mm1.8mm−28% ↓

The penumbras measured for different setups (field size and depth of measurement), are presented in Fig. 2. The left column (Fig. 2(a) for leaf-end and (b) for leaf-side) is from Synergy-S and the right column is from Novalis (Fig. 2c for leaf-end and 2d for leaf-side). For a constant SAD/SCD, the penumbra increase with the measurement depth is more pronounced for larger field sizes due to additional scattering.

  • View full-size image.
  • Figure 2 

    Leaf-end and leaf-side penumbras for different square fields, depth, d=1.5, 5, 10, 15cm for 6MV photon for 100cm SAD and collimator at 0-deg. (a,b) Elekta's Synergy-S, and (c,d) BrainLAB's Novalis.

The effective penumbras as measured for circular fields (from 3cm to 9cm in diameter) are comparable for Novalis' and Synergy-S' μMLC for larger field sizes and larger measurement depths (Fig. 3). The Novalis' μMLC has a consistently smaller penumbra for the circular fields and compared to the Synergy-S' MLC. It is to be noted that the measured penumbra for Novalis (BrainLAB) closely matched with that reported by Clark et al. [4], Chern et al. [5], and García-Garduño et al. [8]. The effective penumbras for these circular fields range from about 4±0.15mm to 7±0.27mm for the Synergy-S μMLC (Fig. 3a) and from about 3±0.12mm to 6±0.23mm for the BrainLAB's μMLC (Fig. 3b).

  • View full-size image.
  • Figure 3 

    Penumbras for different circular fields, depth, d=1.5, 5, 10, 15cm for 6MV photon for 100cm SAD and collimator at 0-deg. (a) Elekta's Synergy-S (on the left panel), and (b) BrainLAB's Novalis (on the right panel).

Tongue-and-groove (T&G) effect 

The tongue-and-groove effects for 6MV photon beam for Synergy-S and Novalis have been presented in Fig. 4. In fact leaves of the Synergy-S Beam Modulator do not have tongue and groove as such; they are machined at a particular angle with respect to the radiation beam, thus making the side of the leaf plane but blocking the beam by overlapping with the adjacent leaves. On the other hand, the Novalis is designed with conventional tongue and grove to minimized interleaf leakage. The maximum T&G effect for Synergy-S is about 25±1% and it is about 23±0.9% for Novalis (20±0.8% for 3mm leaves and 23±0.9% for 5.5mm leaves). It should be noticed that the smaller central leaves (3mm width) produce smaller T&G effects as compared to 4.5mm and 5.5mm leaves (Fig. 4b).

Leakage 

In this study, the leakage for 6MV beam has been quantified as percentage for a delivery of 5000MU (about 5000cGy). The average and maximum leakage for the Synergy-S MLC is 0.9±0.04% and 1.6±0.07%, respectively, (Fig. 5a). The average and maximum leakages of the Novalis are 1.3±0.05% and 2.0±0.08%, respectively. It is to be noted that the leakage through 3mm leaves is larger than through 4.5mm or 5.5mm leaves (Fig. 5b). Thus, Synergy-S Beam Modulator has smaller leakage as compared to that of Novalis M3 μMLC, which is probably due to larger leaf width.

Back to Article Outline

Conclusions 

In this study, the penumbra, tongue-and-groove effect, and transmission through leaves for the μMLC of two linacs (Elekta's Synergy-S and BrainLAB's Novalis) have been investigated. Measurements made using radiographic EDR2 films showed that both the BrainLAB's and the Elekta's μMLC produce comparable effective penumbras for circular fields for larger field sizes and larger measurement depths. The BrainLAB's μMLC has smaller effective penumbra for smaller circular fields (<6cm in diameter), which are typical in radiosurgery and shallower (<5cm) depths. However, for square fields, the Synergy-S appears to produce larger penumbra.

An important finding of this study is the large change of penumbra (17%–34%) while the collimator is rotated by 90° (Fig. 1 and Table 1). Appears that tighter penumbra can be achieved with collimator at 90° as compared to 0° configuration. This can be dosimetrically significant while planning with tighter margin to the tumor. The measurements with collimator rotation suggested the elliptical shape of the Synergy-S' radiation source and roughly circular shape of the Novails' radiation source. The elliptical shape of the sources has important implication on the size of the penumbra (especially for Synergy-S) and may be considered for improving dose distribution.

Back to Article Outline

Disclaimer 

The contents and comments of this manuscript are the sole responsibility of the authors and do not necessarily represent the views of the manufacturer and vendors of the systems involved.

Back to Article Outline

References 

  1. Balog JP, Mackie TR, Wenman DL, Glass M, Fang G, Pearson D. Multileaf collimator interleaf transmission. Med. Phys. 1999;26:176–186
  2. Bortfeld T, Oelfke U, Nill S. What is the optimum leaf width of a multileaf collimator?. Med. Phys. 2000;27:2494–2502
  3. Boyer A, Biggs P, Galvin J, Klein E, LoSasso T, Low D, et al. Basic applications of multileaf collimators. AAPM Report No. 72 Madison, WI: Medical Physics Publishing; 2001;
  4. Clark BG, Teke T, Otto K. Penumbra evaluation of the varian millennium and BrainLAB M3 multileaf collimators. Int. J. Radiat. Oncol. Biol. Phys. 2006;66:S71–S75
  5. Chern SS, Leavitt DD, Jensen RL, Shrieve DC. Is smaller better? Comparison of 3-mm and 5-mm leaf size for stereotactic radiosurgery: a dosimetric study. Int. J. Radiat. Oncol. Biol. Phys. 2006;66:S76–S81
  6. Galvin JM, Smith AR, Lally B. Characterization of a multi-leaf collimator system. Int. J. Radiat. Oncol. Biol. Phys. 1993;25:181–192
  7. Galvin JM, Han K, Cohen R. A comparison of multileaf collimator system and alloy-block field shaping. Int. J. Radiat. Oncol. Biol. Phys. 1998;40:721–731
  8. García-Garduño OA, Celis MA, Lárraga-Gutiérrez JM, Moreno-Jiménez S, Martínez-Dávalos A, Rodríguez-Villafuerte M. Radiation transmission, leakage and beam penumbra measurements of a micro-multileaf collimator using GafChromic EBT film. Appl. Clin. Med. Phys. 2008;9(3):
  9. Huq MS, Das IJ, Steinberg T, Galvin JM. Dosimetric comparison of various multileaf collimators. Phys. Med. Biol. 2002;47:N159–N170
  10. Jordan TJ, Williams PC. The design and performance characteristics of a multileaf collimator. Phys. Med. Biol. 1994;39:231–251
  11. Monk JE, Perks JR, Doughty D, Plowman PN. Comparison of a micro-multileaf collimator with a 5-mmleaf-width collimator for intracranial stereotactic radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 2003;57:1443–1449
  12. Palta JR, Yeung DK, Frouhar V. Dosimetric considerations for a multileaf collimator system. Med. Phys. 1996;23:1219–1224
  13. Pai S, Das I, Dempsey JF, Lam KL, LoSasso TJ, Olch AJ, et al. TG-69: radiographic film for megavoltage beam dosimetry. Med. Phys. 2007;34:2228–2258
  14. Patel I, Glendinning AG, Kirby MC. Dosimetric characteristics of the Elekta Beam Modulator™. Phys. Med. Biol. 2005;50:5479–5492
  15. Wang L, Movsas B, Jacob R, Fourkal E, Chen L, Price R, et al. Stereotactic IMRT for prostate cancer: dosimetric impact of multileaf collimator leaf width in the treatment of prostate cancer with IMRT. J. Appl. Clin. Med. Phys. 2004;5:29–41
  16. Xia P, Geis P, Xing L, Ma C, Findley D, Forster K, et al. Physical characteristics of a miniature multileaf collimator. Med. Phys. 1999;26:65–70

PII: S1120-1797(10)00007-4

doi:10.1016/j.ejmp.2010.01.005

Physica Medica: European Journal of Medical Physics
Volume 27, Issue 1 , Pages 52-57, January 2011

Article Tools

* Image Usage & Resolution